Nearly every child who receives an antipsychotic medicine is first prescribed one of the second-generation, or "atypical" drugs, such as olanzapine and risperidone. However, there has never been evidence that these drugs are more effective than the older, first-generation medications.

Now a study from the University of North Carolina (UNC) at Chapel Hill School of Medicine suggests that molindone, a first-generation drug, is as effective as the newer ones and should be used as a first line of therapy in some children with schizophrenia or schizoaffective disorder.

"People thought the second-generation drugs were superior because they had no side effects. We found that molindone works as well as newer drugs, and in some cases it's safer," said Linmarie Sikich, M.D., associate professor in the department of psychiatry at UNC and lead author of the study, titled the Treatment of Early-Onset Schizophrenia Spectrum Disorders Study.

"The guidelines are going to have to be rewritten because of this study," Sikich said. A report of the study is published Monday in the Sept. 15, 2008 edition of the American Journal of Psychiatry. Co-author Jeffrey Lieberman, M.D., led the earlier CATIE trial at UNC, a landmark study that compared antipsychotic medications in adults. Lieberman is now at Columbia University Medical Center.

The study is the largest head-to-head trial comparing the newer drugs, which became available in the 1990s, to the older ones, which have been around since the 1950s.

Between 2002 and 2006 the study randomly assigned 119 people aged 8 to 19 years to receive either olanzapine (2.5-20 mg/day), risperidone (0.5-6 mg/day), or molindone (10-140 mg/day, plus 1 mg/day of benztropine) over an eight-week period at four sites: UNC; McLean Hospital and Cambridge Health Alliance at Harvard Medical School; Seattle Children's Hospital and the University of Washington; and Case Western Reserve University.

Response rates were similar across the three medications (molindone: 50%; olanzapine: 34%; risperidone: 46%), but the drugs caused very different types of side effects. Both olanzapine and risperidone were associated with significant weight gain and could put young patients at risk of developing heart disease and diabetes. In fact, the National Institute of Mental Health, which sponsored the study, halted recruitment into the olanzapine arm of the study because of the weight gain problem and the resulting increase in cholesterol and glucose levels. There were more reported cases of restlessness with molindone treatment than with either of the two newer treatments.

"Olanzapine should not be a first-line therapy in adolescents," Sikich said.